1058 
Journal of Agricultural Research 
Vol. XXXI, No. 11 
second time on July 27, 1920, and was not bred again prior to the 
beginning of this study. 
Cow 887, a purebred Jersey, was dropped September 5, 1917. 
She was fresh twice, January 11, 1920, ana March 11, 1921, and was 
not bred again prior to the beginning of this study. 
RATIONS 
The same feeding stuffs, mixed in the same proportions, were fed 
throughout the experiment. The quantity fed each animal was so 
adjusted as to maintain approximately constant live weight in 
experiments 221D-885-I and III, experiments 221D-886-I and III, 
experiment 221E-885-II, experiment 221F-874-II, and experiment 
221F-887-II. In the other periods, 221D-885-II, 221D-886-II, 
221E-885-I, 221F-874-I, and 221F-887-I the feed was increased 
so that the animals gained materially in body weight. 
The grain mixture fed was composed of 30 parts wheat bran, 30 
parts ground oats, 30 parts corn meal, and 10 parts old process linseed 
meal, all of good quality. The roughage was good-quality alfalfa, 
grown in Colorado, nicely cured, of uniform green color, containing 
a normal proportion of leaves. 
The details of methods of handling, weighing, sampling and analysis 
of feeds, feces and urine have been published (4, P • 8-9 ). 
For a description of the respiration calorimeter and the minutiae 
of methods for its operation, reference is made to the publications 
of Armsby and Fries. 
DIGESTIBILITY OF THE RATION 
Owing to the impossibility of obtaining attendants, it was imprac¬ 
ticable to collect the dung and urine separately during the digestion 
period. Hence, in order to get an approximate value for the apparent 
digestibility of the ration, the dung and urine were collected separately 
on one day of the experimental period by members of the institute 
staff. On the other days of the digestion period the dung and urine 
were collected together, by means of a conducting apron devised by 
one of the writers, Fries. 
Assuming the relative constancy of composition of the dry matter 
of the dung under the conditions of the experiment, it is possible, by 
making use of the percentage of crude fiber in the dung and urine 
mixture, and in the dung alone, to approximate the apparent digesti¬ 
bility of the various constituents of the feed. The details of this 
computation will be more fully discussed later. 
LIVE WEIGHT 
The animals were weighed daily after the morning feeding and 
before watering. During the 48 hours spent in the respiration 
calorimeter live weights could not be taken. The animals, however, 
were weighed on entering the chamber and again on leaving it. For 
the average weight of the animals during the experimental periods 
the weights on the 10 days immediately preceding the calorimeter 
days, including the last two days of the preliminary period, have been 
used (Table II). 
